Discrimination of frequency glides with superimposed random glides in level

Area of Science:

• Auditory Perception
• Psychoacoustics
• Signal Processing

Background:

• Frequency glides are common in speech and music.
• Previous models suggested detection via excitation level changes on one side of the auditory filter.
• The role of auditory filter characteristics in glide discrimination is not fully understood.

Purpose of the Study:

• To test the hypothesis that frequency glides are detected by monitoring excitation level changes on the low-frequency side of the excitation pattern.
• To measure thresholds for detecting increases in frequency glide extent.
• To investigate the influence of various auditory parameters on glide discrimination.

Main Methods:

• Measured detection thresholds for frequency glide extent across different standard transition spans.
• Roved center frequencies and varied sound levels (fixed or randomly changing) to prevent cueing.
• Expressed thresholds as a proportion of the auditory filter's equivalent rectangular bandwidth (ERB).

Main Results:

• Performance exceeded predictions based on monitoring one side of the excitation pattern or using start/endpoint frequencies.
• Thresholds showed little dependence on center frequency (0.5-6 kHz), glide duration (50-400 ms), or glide direction.
• Thresholds increased significantly when the standard transition span exceeded 0.5 ERB.

Conclusions:

• Frequency glide extent changes can be discriminated independently of monitoring just one side of the excitation pattern.
• Discrimination is not solely reliant on phase-locking information.
• The findings suggest a more complex mechanism for processing frequency glide extent in auditory perception.